Apparatus and Method for Privacy Protection in Association Rule Mining

1. An apparatus, comprising: a fake transaction inserter that generates fake transactions of a predetermined number each having a predetermined length and inserts the fake transactions between a plurality of transactions comprised in an original data set to generate a first virtual data set; and a distortion transaction generator that protects the privacy of the original data set by generating a second virtual data set by converting data of the transaction of the first virtual data set with a predetermined probability, wherein the distortion transaction generator generates the transaction of the second virtual data set by using a result of an exclusive OR function having the conversion variable and the transaction of the first virtual data set corresponding to the conversion variable as an independent variables.

2. The apparatus of claim 1 , wherein the predetermined length is a uniform distribution random variable in which a predetermined number of length variables are distributed within a predetermined range with the same probability.

3. The apparatus of claim 1 , wherein an average of the predetermined length is same with an average of a length of the transactions of the original data set.

4. The apparatus of claim 1 , wherein the predetermined number is the uniform distribution random variable having an average length and a range determined by using a predetermined security parameter.

5. The apparatus of claim 1 , wherein the distortion transaction generator further includes a function to generate conversion variables, having a value of 0 or 1 as the predetermined probability, of the same number as the number of the transactions of the first virtual data set.

6. The apparatus of claim 5 , wherein the conversion variable is a Bernoulli function value having the predetermined probability as an independent variable.

7. A method, comprising: receiving an original data set from a server and calculating an average length of a plurality of transactions comprised in the original data set by a fake transaction inserter; generating fake transactions of a predetermined number each having a predetermined length and insert the fake transactions between the plurality of transactions to generate a first virtual data set by the fake transaction inserter; and protecting the privacy of the original data set by generating a second virtual data set by converting data of the transaction of the first virtual data set with a predetermined probability by a distortion transaction generator, wherein the generating the second virtual data set generates the transaction of the second virtual data set by using a result of an exclusive OR function having the conversion variable and the transaction of the first virtual data set corresponding to the conversion variable as an independent variables.

8. The method of claim 7 , wherein the predetermined length is a uniform distribution random variable in which a predetermined number of length variables are distributed within a predetermined range with the same probability.

9. The method of claim 7 , wherein an average of the predetermined length is same with an average of a length of the transactions of the original data set.

10. The method of claim 7 , wherein the predetermined number is the uniform distribution random variable having an average length and a range determined by using a predetermined security parameter.

11. The method of claim 7 , wherein the generating the second virtual data set further includes generating conversion variables, having a value of 0 or 1 as the predetermined probability, of the same number as the number of the transactions of the first virtual data set.

12. The method of claim 11 , wherein the conversion variable is a Bernoulli function value having the predetermined probability as an independent variable.